This cohort study examines data from general practice, hospital, and mortality records to determine whether individuals with cataract, age-related macular degeneration, or glaucoma have higher risks of falls or fractures than those without.
Key Points
Question
Do people with cataract, age-related macular degeneration (AMD), or glaucoma have higher risks of falls or fractures?
Findings
In this cohort study including 3 434 196 adults, we found an increased risk of falls in those with cataract, AMD, and glaucoma. For fractures, there was also an increased risk for those with cataract, AMD, and glaucoma.
Meaning
The results of this study support recognition that people with 1 or more of these eye diseases are at increased risk of falls or fractures.
Abstract
Importance
Three leading disease causes of age-related visual loss are cataract, age-related macular degeneration (AMD), and glaucoma. Although all 3 eye diseases have been implicated with falls and fracture risk, evidence is mixed, with the contribution of different eye diseases being uncertain.
Objective
To examine whether people with cataract, AMD, or glaucoma have higher risks of falls or fractures than those without.
Design, Setting, and Participants
This cohort study was a population-based study in England using routinely collected electronic health records from the Clinical Practice Research Datalink (CPRD) GOLD and Aurum primary care databases with linked hospitalization and mortality records from 2007 to 2020. Participants were people with cataract, AMD, or glaucoma matched to comparators (1:5) by age, sex, and general practice. Data were analyzed from May 2021 to June 2023.
Exposures
For each eye disease, we estimated the risk of falls or fractures using separate multivariable Cox proportional hazards regression models.
Main Outcomes
Two primary outcomes were incident falls and incident fractures derived from general practice, hospital, and mortality records. Secondary outcomes were incident fractures of specific body sites.
Results
A total of 410 476 people with cataract, 75 622 with AMD, and 90 177 with glaucoma were matched (1:5) to 2 034 194 (no cataract), 375 548 (no AMD), and 448 179 (no glaucoma) comparators. The mean (SD) age was 73.8 (11.0) years, 79.4 (9.4) years, and 69.8 (13.1) years for participants with cataract, AMD, or glaucoma, respectively. Compared with comparators, there was an increased risk of falls in those with cataract (adjusted hazard ratio [HR], 1.36; 95% CI, 1.35-1.38), AMD (HR, 1.25; 95% CI, 1.23-1.27), and glaucoma (HR, 1.38; 95% CI, 1.35-1.41). Likewise for fractures, there were increased risks in all eye diseases, with an HR of 1.28 (95% CI, 1.27-1.30) in the cataract cohort, an HR of 1.18 (95% CI, 1.15-1.21) for AMD, and an HR of 1.31 (95% CI, 1.27-1.35) for glaucoma. Site-specific fracture analyses revealed increases in almost all body sites (including hip, spine, forearm, skull or facial bones, pelvis, ribs or sternum, and lower leg fractures) compared with matched comparators.
Conclusions and Relevance
The results of this study support recognition that people with 1 or more of these eye diseases are at increased risk of both falls and fractures. They may benefit from improved advice, access, and referrals to falls prevention services.
Introduction
Three leading disease causes of age-related visual loss are cataract, age-related macular degeneration (AMD), and glaucoma, affecting more than 500 million people worldwide.1 Visual loss increases morbidity and mortality, including physical injuries, disability, poor cognition, and decreased mental health, leading to a reduction in activities of daily living and a loss of independence.2,3 The majority of these eye diseases are preventable or treatable, which may in turn reduce the risk of falls and related injuries, carrying important resource implications for global health and individual health systems dealing with an aging society.2,4,5 Falls are a major global health concern, particularly as the second leading cause of unintentional deaths due to injury worldwide.6 Annually, there are more than 650 000 deaths due to falls and more than 170 million falls resulting in short-term or long-term disability.5 This translates to an estimated cost of $23.3 billion annually in the United States and $1.6 billion in the United Kingdom.7
Poor vision is one of many risk factors for falls, but links to specific eye diseases remain inadequately defined. Visual function is vital for avoiding falls, with even relatively mild impairments in visual information affecting balance, posture, and gait.2,8 Yet in early stages of eye disease, patients are often asymptomatic and unaware of visual impairment.9 Both cataract and AMD mostly start affecting a single eye with a gradual onset in visual loss.8 In glaucoma, there is often insidious peripheral visual field loss, but the brain perceptually compensates for the missing areas by artificially completing the visual field.10 Although all 3 eye diseases have been implicated with falls and fracture risk, evidence is mixed, with current findings mainly derived from cross-sectional observations and having limited adjustment for established risk factors contributing to fall and fracture risk.11,12,13,14 Though smaller studies have reported an increased risk of falls and fractures, both the magnitude and contribution of each individual eye disease to these risks remain uncertain.8,15,16,17,18 Therefore, this study sought to determine the association of 3 leading age-related eye diseases with falls and fractures, adjusting for influential risk factors. The overarching aim was to investigate whether individuals with cataract, AMD, or glaucoma are at higher relative risk of falls or fractures compared with individuals without these eye diseases.
Methods
Study Design and Data Sources
This study was a population-based retrospective cohort study using the Clinical Practice Research Datalink (CPRD) GOLD and Aurum UK primary care databases.19,20 These contain anonymized longitudinal medical records from 2 of the most widely used clinical information systems in the United Kingdom, named Vision (GOLD) and EMIS Web (Aurum). The study was approved by CPRD’s independent scientific advisory committee and the Medicines and Healthcare products Regulatory Agency independent scientific advisory committee. Given the retrospective use of anonymized data, no informed consent was required. This study was conducted according to the guidelines of the Declaration of Helsinki, and the reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.
Setting
In the United Kingdom, the National Health Service is free at the point of care, with primary care functioning as gatekeepers of access to specialty care. Electronic records are adopted across all primary care practices, and data are collected daily from voluntarily enrolled practices by CPRD, which is then integrated with existing records and subjected to multiple quality checks.21,22 The data are nationally representative in terms of age, sex, and race and ethnicity, collated from more than 2200 primary care practices and including 18 million active patients across the United Kingdom, approximately 25% of the UK population.19,20,23 The patient-level data include detailed information on demographics, clinical events, prescriptions, and specialist referrals.
CPRD GOLD and Aurum data sets were combined for analyses as per previous studies.23,24 As there was a small overlap of practices that migrated clinical information software over time, a bridging file was used to drop practices who migrated from GOLD to Aurum to avoid double counting. Patient records were only included if deemed of acceptable quality for research (via a CPRD quality metric in GOLD). All included records were derived from practices based in England and linked at the patient level to Hospital Episode Statistics for hospitalization data, to the Office for National Statistics for mortality data, and by small area to the Index of Multiple Deprivation (IMD) 2015 stratified as quintiles.
Participants and Cohort Delineation
Our study population included 3 separate cohorts (Figure 1) of adults 18 years and older, with cases defined as having a recorded diagnosis of cataract, AMD, or glaucoma (allowing concurrent eye disease within each cohort). These were identified from each database using Read codes (and additional SNOMED Clinical Terms EMIS-specific codes for CPRD Aurum) between April 1, 2007, and March 31, 2020, with previous studies demonstrating high validity in eye disease coding (eTable 1 in Supplement 1 contains lists of codes).25,26 However, we were unable to ascertain whether the eye disease was a monocular or binocular diagnosis.
Figure 1. Study Flowchart Showing Cohorts Identified From the UK Clinical Practice Research Datalink (CPRD) GOLD and Aurum.
AMD indicates age-related macular degeneration.
Each case was matched with up to 5 corresponding comparators on age, sex, and general practice using incidence density sampling (99% cases were successfully matched with 5 comparators), with no recorded diagnosis of primary eye disease (ie, comparators were allowed to have a past diagnosis of the 2 other eye diseases). Study entry was defined as the first recorded eye-disease diagnosis date (at any point within the follow-up period but with at least 1-year registration within a general practice), with the end of follow-up defined as the earliest of death date, study end date (March 31, 2020), date of deregistration from a practice, or last data collection by the practice.
Outcomes and Covariates
The 2 primary outcomes were rates of incident falls and incident fractures. These were identified using a predefined list of Read codes from primary care and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, codes from linked secondary care records.27 Diagnostic codes for pathological fractures were excluded to focus on trauma-related fractures. A secondary analysis examining the rate of site-specific fractures was also performed. This included all body sites of coded fractures with classifications guided by the National Institute for Health and Care Excellence guidelines on osteoporosis and previous studies.28,29,30
Covariates included racial and ethnic group as reported by patients (Asian, Black, White, other, and unknown), patient-level deprivation score, Charlson Comorbidity Index, smoking history, and heavy alcohol use.31,32,33 The deprivation score was formed by reversing CPRD quintiles to be consistent with definitions of IMD quintiles (1 = most deprived), providing an area-level measure of approximately 1500 people and a composite score across 7 domains, such as income, employment, education, etc. Any history of other eye disease, osteoporosis, fractures, and falls before the index date was also included.
We examined specific medication groups known to increase the risk of falls and fracture risk, including benzodiazepines, antidepressants, cardiovascular drugs (including antihypertensives and α-blockers), antidiabetes drugs (with a separate category for insulin), anticholinergics (only those with anticholinergic burden 2 or 3 were included, from Richardson et al34), and systemic steroids.35,36 These medication group covariates were measured before the index diagnosis date, including any history within 12 months prior, and treated as binary variables.
Statistical Analyses
Incident rates are presented in age-standardized rates per 100 000 population using the Global Burden of Diseases population structure.37 Multivariable Cox proportional hazards regression was used to estimate the risk of incident falls and incident fractures (constituting events within separate models) in cases compared with comparators and to examine associations with covariates. To account for our matched cohort design, all Cox models were stratified by matched sets. The proportional hazards assumption was assessed visually via log-log plots. Missing data for IMD and race and ethnicity were coded by creating a missing category. Missing smoking status was imputed with the multivariate imputation by a chained-equations algorithm using 10 complete data sets.
Sensitivity analyses were performed using propensity scores to account for covariate imbalance between the cohorts within further Cox proportional hazard models.38 Variables related to each outcome at P < .05 were selected for inclusion in propensity weights, identified through various regression analyses with each covariate.39 Inverse-weighted probability models included more than 50 confounders covering demographics, long-term conditions, falls-risk–inducing medications, eye medications, and interaction terms for composite measures such as Charlson Comorbidity Index. Further sensitivity analyses were also performed for populations with a single eye disease only (ie, without concurrent history of the other eye diseases) to assess effects on outcomes. We also performed an additional analysis by separating populations for CPRD GOLD and Aurum to observe any differences between the databases (eTable 7 in Supplement 1).
Data were analyzed from May 2021 to June 2023. All analyses were performed using Stata version 16.1 (StataCorp). All P values were 2-sided, and there were no adjustments for multiple analyses.
Results
Baseline Characteristics
The delineation of each study cohort is specified in Figure 1 with baseline characteristics described in Table 1. All cohorts mainly consisted of older adults, with a mean (SD) age of 74.3 (11.5) years (mean [SD] ages of 73.8 [11.0] years, 79.4 [9.4] years, and 69.8 [13.1] years for cataract, AMD, and glaucoma, respectively) and a slightly higher proportion of females (57.2%, 62.2%, and 51.9% respectively). The cataract cohort included 410 476 cases to 2 034 194 comparators and a median (IQR) follow-up of 4.02 years (2.07-6.86 years) in cases vs 3.91 years (1.97-6.86 years) in comparators. The AMD cohort had 75 622 cases to 375 548 comparators and a median (IQR) follow-up of 3.76 years (1.91-6.49 years) in cases vs 3.61 years (1.76-6.45 years) in comparators. The glaucoma cohort consisted of 90 177 cases to 448 179 comparators and a median (IQR) follow-up of 4.46 years (2.29-7.55 years) in cases vs 4.24 (2.10-7.36 years) in controls.
Table 1. Baseline Characteristics for Matched Study Cohorts.
| Characteristic | Cataract, No. (%) | AMD, No. (%) | Glaucoma, No. (%) | |||
|---|---|---|---|---|---|---|
| Cases (n = 410 476) | Controls (n = 2 034 194) | Cases (n = 75 622) | Controls (n = 375 548) | Cases (n = 90 177) | Controls (n = 448 179) | |
| Follow-up, median (IQR), y | 4.0 (2.1-6.9) | 4.0 (2.0-6.9) | 3.8 (1.9-6.5) | 3.6 (1.8-6.5) | 4.5 (2.3-7.6) | 4.2 (2.1-7.4) |
| Age, mean (SD) | 73.8 (11.0) | 73.8 (11.0) | 79.4 (9.6) | 79.3 (9.5) | 69.8 (13.1) | 69.8 (13.1) |
| Age category, y | ||||||
| 18-44 | 6703 (1.6) | 33 352 (1.6) | 311 (0.4) | 1553 (0.4) | 3615 (4.0) | 17 968 (4.0) |
| 45-64 | 65 949 (16.1) | 328 174 (16.1) | 5136 (6.8) | 25 597 (6.8) | 24 261 (26.9) | 120 662 (26.9) |
| ≥65 | 337 824 (82.3) | 1 672 668 (82.2) | 70 175 (92.8) | 348 398 (92.8) | 62 301 (69.1) | 309 549 (69.1) |
| Deaths (all causes) | 78 570 (19.1) | 362 517 (17.8) | 19 601 (25.9) | 97 822 (26.1) | 13 125 (14.6) | 68 063 (15.2) |
| Sex | ||||||
| Male | 176 018 (42.9) | 871 002 (42.8) | 28 699 (38.0) | 142 113 (37.8) | 43 425 (48.2) | 215 681 (48.1) |
| Female | 234 458 (57.1) | 1 163 192 (57.2) | 46 923 (62.1) | 233 435 (62.2) | 46 752 (51.8) | 232 498 (51.9) |
| Race and ethnicitya | ||||||
| Asian | 16 377 (4.0) | 36 130 (1.8) | 1103 (1.5) | 4999 (1.3) | 3370 (3.7) | 10 799 (2.4) |
| Black | 10 227 (2.5) | 28 789 (1.4) | 400 (0.5) | 3569 (1.0) | 4151 (4.6) | 8544 (1.9) |
| White | 362 322 (88.3) | 1 631 553 (80.2) | 71 009 (93.9) | 325 072 (86.6) | 75 233 (83.4) | 346 985 (77.4) |
| Other | 2660 (0.7) | 20 887 (1.0) | 290 (0.4) | 2745 (0.7) | 697 (0.8) | 5456 (1.2) |
| Unknown | 18 890 (4.6) | 316 835 (15.6) | 2820 (3.7) | 39 163 (10.4) | 6726 (7.5) | 76 395 (17.1) |
| IMD | ||||||
| 1 (Most deprived) | 16 (15.8) | 305 975 (15.0) | 11 061 (14.6) | 54 872 (14.6) | 13 723 (15.2) | 67 583 (15.1) |
| 2 | 74 740 (18.2) | 367 573 (18.1) | 13 306 (17.6) | 65 705 (17.5) | 16 150 (17.9) | 80 710 (18.0) |
| 3 | 82 422 (20.1) | 413 527 (20.3) | 15 276 (20.2) | 75 869 (20.2) | 17 774 (19.7) | 90 730 (20.2) |
| 4 | 88 259 (21.5) | 444 586 (21.9) | 16 933 (22.4) | 84 500 (22.5) | 19 803 (22.0) | 97 602 (21.8) |
| 5 (Least deprived) | 100 036 (24.4) | 500 431 (24.6) | 18 999 (25.1) | 94 200 (25.1) | 22 679 (25.2) | 111 189 (24.8) |
| Unknown | 246 (0.1) | 2102 (0.1) | 47 (0.1) | 402 (0.1) | 48 (0.1) | 365 (0.1) |
| CCI category | ||||||
| None | 177 983 (43.4) | 1 827 218 (89.8) | 27 541 (36.4) | 304 401 (81.1) | 48 675 (54.0) | 396 212 (88.4) |
| Mild (1-2) | 142 315 (34.7) | 124 426 (6.1) | 27 307 (36.1) | 40 436 (10.8) | 27 137 (30.1) | 30 439 (6.8) |
| Moderate (3-4) | 64 973 (15.8) | 60 543 (3.0) | 14 639 (19.4) | 21 950 (5.8) | 10 463 (11.6) | 15 280 (3.4) |
| Severe (≥5) | 25 205 (6.1) | 22 007 (1.1) | 6135 (8.1) | 8761 (2.3) | 3902 (4.3) | 6248 (1.4) |
| Previous eye disease | ||||||
| Cataract | NA | NA | 49 405 (65.3) | 148 755 (39.6) | 39 446 (43.7) | 109 983 (24.5) |
| AMD | 41 039 (10.0) | 91 816 (4.5) | NA | NA | 6481 (7.2) | 20 666 (4.6) |
| Glaucoma | 50 191 (12.2) | 112 015 (5.5) | 9796 (13.0) | 33 364 (8.9) | NA | NA |
| Medical historyb | ||||||
| Cardiovascular disease | 107 768 (26.3) | 332 275 (16.3) | 23 450 (31.0) | 85 065 (22.7) | 17 176 (19.1) | 68 383 (15.3) |
| Hypertension | 171 979 (41.9) | 159 275 (7.8) | 35 667 (47.2) | 54 921 (14.6) | 32 218 (35.7) | 38 607 (8.6) |
| Type 1 diabetes | 13 149 (3.2) | 7927 (0.4) | 2325 (3.1) | 3385 (0.9) | 1954 (2.2) | 2515 (0.6) |
| Type 2 diabetes | 84 438 (20.6) | 53 253 (2.6) | 13 916 (18.4) | 19 924 (5.3) | 13 226 (14.7) | 17 342 (3.9) |
| Asthma/COPD | 19 532 (1.0) | 58 932 (14.4) | 8260 (2.2) | 8189 (10.8) | 6940 (1.6) | 9259 (10.3) |
| Neurological condition | 3132 (0.2) | 9687 (2.4) | 1279 (0.3) | 1268 (1.7) | 1052 (0.2) | 1854 (2.1) |
| Liver disease | 677 (0.0) | 2783 (0.7) | 218 (0.1) | 322 (0.4) | 266 (0.1) | 377 (0.4) |
| Kidney disease | 22 532 (1.1) | 53 606 (13.1) | 11 747 (3.1) | 9744 (12.9) | 7655 (1.7) | 6977 (7.7) |
| Thyroid disease | 11 547 (0.6) | 29 534 (7.2) | 4597 (1.2) | 4457 (5.9) | 3331 (0.7) | 4499 (5.0) |
| Connective tissue disease | 15 125 (0.7) | 40 633 (9.9) | 5789 (1.5) | 5584 (7.4) | 4710 (1.1) | 6213 (6.9) |
| Cancer | 21 718 (1.1) | 51 756 (12.6) | 9594 (2.6) | 8543 (11.3) | 6777 (1.5) | 7898 (8.8) |
| Mental health condition | 34 608 (1.7) | 95 158 (23.2) | 12 586 (3.4) | 12 609 (16.7) | 10 109 (2.3) | 17 601 (19.5) |
| Dementia | 2706 (0.1) | 6342 (1.6) | 2794 (0.7) | 1387 (1.8) | 1424 (0.3) | 1024 (1.1) |
| Osteoporosis | 41 810 (10.2) | 40 356 (2.0) | 10 375 (13.7) | 16 829 (4.5) | 6716 (7.5) | 10 374 (2.3) |
| Current smoker | 207 660 (50.6) | 901 758 (44.3) | 39 467 (52.2) | 153 449 (40.9) | 42 365 (47.0) | 200 560 (44.8) |
| Heavy alcohol use | 27 176 (6.6) | 17 735 (0.9) | 4107 (5.4) | 4847 (1.3) | 5550 (6.15) | 4928 (1.1) |
| Fall history | 88 742 (21.6) | 138 892 (6.8) | 20 922 (27.7) | 47 497 (12.7) | 15 331 (17.0) | 31 667 (7.1) |
| Fracture history | 44 827 (10.9) | 110 346 (5.4) | 9931 (13.1) | 32 152 (8.6) | 7909 (8.8) | 22 277 (5.0) |
| Medication use | ||||||
| Antidepressants | 86 303 (21.0) | 95 502 (4.7) | 14 827 (19.6) | 20 216 (5.4) | 14 950 (16.6) | 15 047 (3.4) |
| Benzodiazepines | 25 906 (6.3) | 20 725 (1.0) | 5506 (7.3) | 7121 (1.9) | 4745 (5.3) | 5161 (1.2) |
| Antihypertensives | 269 556 (65.7) | 235 195 (11.6) | 54 024 (71.4) | 81 673 (21.8) | 48 741 (54.1) | 59 648 (13.3) |
| Antidiabetes drugs | 67 377 (16.4) | 39 851 (2.0) | 10 360 (13.7) | 14 809 (3.9) | 10 189 (11.3) | 13 362 (3.0) |
| Insulin | 21 003 (5.1) | 8932 (0.4) | 2870 (3.8) | 3801 (1.0) | 3326 (3.7) | 3750 (0.8) |
| Systemic steroids | 46 593 (11.4) | 32 189 (1.6) | 8217 (10.9) | 11 721 (3.1) | 7376 (8.2) | 8665 (1.9) |
| High ACB drugs (ACB 2 or 3) | 76 843 (18.7) | 59 042 (2.9) | 14 689 (19.4) | 20 722 (5.5) | 14 352 (15.9) | 15 502 (3.5) |
Abbreviations: ACB, anticholinergic burden; AMD, age-related macular degeneration; CCI, Charlson Comorbidity Index; COPD, chronic obstructive pulmonary disease; IMD, Index of Multiple Deprivation 2015; NA, not applicable.
The race and ethnicity category other includes patients who reported races or ethnicities that were not Asian, Black, or White.
Medical history has been grouped for ease of presentation. Cardiovascular disease includes coronary heart disease, heart failure, peripheral vascular disease, and cerebrovascular disease. Neurological condition includes epilepsy, multiple sclerosis, and Parkinson disease. Mental health condition includes anxiety, depression, eating disorder, bipolar disorder, and schizophrenia. Connective tissue disease includes rheumatoid arthritis, psoriatic arthritis, polyarthritis, and spondyloarthropathies.
At baseline, all eye disease populations had poorer health and a higher level of comorbidity, including a greater prevalence of all multiple long-term conditions compared with comparators in both physical health (eg, cardiovascular disease, respiratory conditions, osteoporosis) and mental health diagnoses (eg, depression, bipolar, dementia). This finding was also reflected in medication usage, with notably higher proportions of cases taking benzodiazepines, antidepressants, antihypertensives, antidiabetes medications, systemic steroids, and medications with a high anticholinergic burden. A higher proportion of people with eye disease had a history of both falls (approximately 3-fold for cataract and 2-fold for AMD and glaucoma) and fractures. People with eye disease were also more likely to have a history of the other 2 eye diseases compared with comparators.
Incident Falls and Incident Fractures
During the study period, there was an increased incidence of both falls and fractures experienced by people with eye disease compared with their matched comparators. eTable 2 in Supplement 1 shows the proportions and crude rates for the primary and secondary outcomes, and eTable 3 in Supplement 1 shows the age-standardized incidence rates. Overall, a greater proportion of people with eye disease compared with comparators experienced falls (cataract, 29.7% vs 13.9%; AMD, 37.1% vs 20.7%; glaucoma, 25.0% vs 12.8%) and fractures (cataract, 14.4% vs 8.2%; AMD, 17.8% vs 11.6%; glaucoma, 12.2% vs 7.3%). The age-standardized incidence rates per 100 000 person-years for falls were 2217.5 (95% CI, 2144.5-2296.1) cases for individuals with cataract compared with 625.0 (95% CI, 611.2-639.6) in comparators, 2551.4 (95% CI, 2246.8-2956.6) in those with AMD compared with 848.1 (95% CI, 788.7-927.2) in comparators, and 1802.0 (95% CI, 1708.8-1903.8) for glaucoma compared with 621.3 (95% CI, 601.0-643.3) in comparators.
Table 2 and Table 3 show the results of the multivariable Cox regression analysis for falls and fractures, with Figure 2 illustrating a comparison of the adjusted hazard ratios (HRs) for each eye disease cohort. Overall, there was an increased risk of falls in those with cataract (HR, 1.36; 95% CI, 1.35-1.38), AMD (HR, 1.25; 95% CI, 1.23-1.27), and glaucoma (HR, 1.38; 92% CI, 1.35-1.41) compared with matched comparators. Likewise for fractures, there also was an increased risk in all eye diseases, with an HR of 1.28 (95% CI, 1.27-1.30) in the cataract cohort, HR of 1.18 (95% CI, 1.15-1.21) for AMD, and HR of 1.31 (95% CI, 1.27-1.35) for glaucoma. As observed, HRs were slightly higher for glaucoma and cataract compared with AMD for both falls and fractures.
Table 2. Multivariable Hazard Ratios for Incident Falls With Adjustment for Covariates.
| Covariate | HR (95% CI) | ||
|---|---|---|---|
| Cataract | AMD | Glaucoma | |
| Case | 1.36 (1.35-1.38) | 1.25 (1.23-1.27) | 1.38 (1.36-1.41) |
| Agea | NA | NA | NA |
| Sexa | NA | NA | NA |
| Raceb | |||
| Asian | 0.86 (0.83-0.88) | 0.82 (0.76-0.89) | 0.80 (0.75-0.86) |
| Black | 0.59 (0.57-0.62) | 0.59 (0.54-0.65) | 0.60 (0.56-0.65) |
| Other | 0.73 (0.69-0.77) | 0.66 (0.59-0.74) | 0.70 (0.63-0.78) |
| Unknown | 0.23 (0.23-0.24) | 0.27 (0.26-0.29) | 0.25 (0.24-0.26) |
| IMDb | |||
| 2 | 0.94 (0.93-0.96) | 0.96 (0.93-0.99) | 0.94 (0.91-0.97) |
| 3 | 0.90 (0.89-0.92) | 0.92 (0.90-0.95) | 0.88 (0.85-0.91) |
| 4 | 0.88 (0.87-0.90) | 0.91 (0.88-0.94) | 0.85 (0.82-0.88) |
| 5 (Least deprived) | 0.86 (0.85-0.88) | 0.88 (0.86-0.91) | 0.84 (0.81-0.87) |
| Unknown | 0.90 (0.78-1.05) | 0.78 (0.59-1.05) | 1.02 (0.72-1.45) |
| Previous eye disease | |||
| Cataract | NA | 1.25 (1.23-1.27) | 1.31 (1.28-1.33) |
| AMD | 1.07 (1.06-1.09) | NA | 1.05 (1.02-1.08) |
| Glaucoma | 1.16 (1.15-1.18) | 1.12 (1.10-1.15) | NA |
| CCI scoreb | |||
| 1 | 1.40 (1.37-1.42) | 1.30 (1.26-1.34) | 1.34 (1.30-1.39) |
| 2 | 1.28 (1.26-1.29) | 1.18 (1.15-1.22) | 1.22 (1.18-1.26) |
| 3 | 1.37 (1.34-1.39) | 1.30 (1.26-1.34) | 1.33 (1.28-1.39) |
| 4 | 1.37 (1.34-1.40) | 1.30 (1.25-1.35) | 1.33 (1.26-1.39) |
| 5 | 1.45 (1.41-1.49) | 1.34 (1.28-1.41) | 1.34 (1.26-1.43) |
| 6 | 1.47 (1.42-1.53) | 1.43 (1.34-1.53) | 1.54 (1.42-1.67) |
| 7 | 1.60 (1.51-1.70) | 1.66 (1.50-1.84) | 1.38 (1.21-1.56) |
| 8 | 1.72 (1.57-1.90) | 1.65 (1.41-1.93) | 2.16 (1.78-2.61) |
| 9 | 1.80 (1.57-2.08) | 1.68 (1.31-2.15) | 2.46 (1.80-3.36) |
| 10 | 1.97 (1.53-2.54) | 2.02 (1.30-3.14) | 2.10 (1.29-3.41) |
| 11 | 2.39 (1.64-3.48) | 0.92 (0.46-1.86) | 1.57 (0.77-3.22) |
| ≥12 | 1.84 (1.16-2.91) | 1.28 (0.58-2.79) | 1.84 (1.17-2.91) |
| Medical history | |||
| Current smoker | 1.07 (1.04-1.10) | 1.04 (1.00-1.10) | 1.00 (0.93-1.08) |
| Heavy alcohol use | 1.42 (1.38-1.45) | 1.24 (1.18-1.31) | 1.40 (1.33-1.48) |
| Osteoporosis | 1.50 (1.48-1.52) | 1.46 (1.42-1.50) | 1.48 (1.43-1.53) |
| Fall history | 1.80 (1.78-1.82) | 1.75 (1.72-1.79) | 1.83 (1.79-1.88) |
| Medication use | |||
| Antidepressants | 1.27 (1.25-1.28) | 1.32 (1.29-1.36) | 1.38 (1.33-1.43) |
| Benzodiazepines | 1.20 (1.18-1.23) | 1.17 (1.13-1.22) | 1.12 (1.07-1.17) |
| Antihypertensives | 1.31 (1.30-1.32) | 1.28 (1.26-1.31) | 1.23 (1.20-1.26) |
| Antidiabetes drugs | 1.18 (1.16-1.20) | 1.12 (1.09-1.15) | 1.17 (1.13-1.22) |
| Insulin | 1.34 (1.30-1.38) | 1.25 (1.18-1.32) | 1.37 (1.29-1.46) |
| ACB-2 drugs | 1.58 (1.51-1.66) | 1.47 (1.35-1.61) | 1.78 (1.61-1.98) |
| ACB-3 drugs | 1.19 (1.17-1.21) | 1.11 (1.08-1.15) | 1.15 (1.11-1.19) |
| Systemic steroids | 1.25 (1.23-1.27) | 1.17 (1.13-1.21) | 1.24 (1.19-1.29) |
Abbreviations: ACB, anticholinergic burden; AMD, age-related macular degeneration; CCI, Charlson Comorbidity Index; HR, hazard ratio; IMD, index of multiple deprivation; NA, not applicable.
As cases and controls were matched on age and sex, there were no differences between the 2 groups in the Cox models.
Race categories were compared with White, IMD compared with 1 (most deprived), and CCI score compared with 0.
Table 3. Multivariable Hazard Ratios for Incident Fractures With Adjustment for Covariates.
| Covariate | HR (95% CI) | ||
|---|---|---|---|
| Cataract | AMD | Glaucoma | |
| Case | 1.28 (1.27-1.30) | 1.18 (1.15-1.21) | 1.31 (1.27-1.35) |
| Agea | NA | NA | NA |
| Sexa | NA | NA | NA |
| Raceb | |||
| Asian | 0.74 (0.71-0.78) | 0.87 (0.78-0.97) | 0.67 (0.61-0.73) |
| Black | 0.34 (0.32-0.36) | 0.35 (0.29-0.41) | 0.38 (0.34-0.43) |
| Other | 0.71 (0.67-0.76) | 0.58 (0.50-0.68) | 0.63 (0.55-0.73) |
| Unknown | 0.20 (0.20-0.21) | 0.22 (0.20-0.23) | 0.21 (0.20-0.23) |
| IMDb | |||
| 2 | 0.96 (0.94-0.98) | 0.95 (0.91-0.98) | 0.96 (0.92-1.00) |
| 3 | 0.91 (0.90-0.93) | 0.94 (0.90-0.97) | 0.91 (0.87-0.95) |
| 4 | 0.89 (0.87-0.91) | 0.92 (0.88-0.96) | 0.86 (0.82-0.90) |
| 5 (Least deprived) | 0.88 (0.86-0.90) | 0.90 (0.86-0.94) | 0.85 (0.81-0.89) |
| Unknown | 0.94 (0.78-1.14) | 0.81 (0.56-1.18) | 0.82 (0.51-1.30) |
| Previous eye disease | |||
| Cataract | NA | 1.11 (1.08-1.13) | 1.20 (1.17-1.23) |
| AMD | 0.99 (0.98-1.01) | NA | 1.00 (0.96-1.04) |
| Glaucoma | 1.07 (1.05-1.08) | 1.03 (1.00-1.06) | NA |
| CCI scoreb | |||
| 1 | 1.23 (1.21-1.26) | 1.18 (1.13-1.22) | 1.22 (1.16-1.27) |
| 2 | 1.15 (1.13-1.17) | 1.07 (1.03-1.11) | 1.07 (1.03-1.12) |
| 3 | 1.21 (1.18-1.24) | 1.14 (1.09-1.20) | 1.13 (1.07-1.19) |
| 4 | 1.22 (1.19-1.26) | 1.18 (1.12-1.24) | 1.16 (1.08-1.24) |
| 5 | 1.23 (1.18-1.28) | 1.11 (1.03-1.19) | 1.09 (1.00-1.20) |
| 6 | 1.28 (1.21-1.35) | 1.25 (1.13-1.37) | 1.20 (1.06-1.35) |
| 7 | 1.38 (1.27-1.51) | 1.32 (1.14-1.52) | 1.13 (0.94-1.36) |
| 8 | 1.60 (1.40-1.82) | 1.36 (1.08-1.71) | 1.55 (1.16-2.05) |
| 9 | 1.48 (1.22-1.80) | 1.66 (1.15-2.37) | 2.11 (1.35-3.28) |
| 10 | 2.36 (1.71-3.28) | 1.41 (0.74-2.70) | 1.81 (0.89-3.66) |
| 11 | 2.03 (1.21-3.40) | 1.52 (0.64-3.61) | 1.32 (0.48-3.62) |
| ≥12 | 1.13 (0.56-2.26) | 0.64 (0.17-2.47) | 1.13 (0.56-2.26) |
| Medical history | |||
| Current smoker | 1.07 (1.04-1.10) | 1.02 (0.96-1.10) | 1.03 (0.94-1.14) |
| Heavy alcohol use | 1.52 (1.47-1.57) | 1.25 (1.17-1.34) | 1.44 (1.34-1.55) |
| Osteoporosis | 1.52 (1.49-1.55) | 1.49 (1.44-1.54) | 1.57 (1.50-1.65) |
| Fracture history | 1.97 (1.93-2.00) | 1.80 (1.74-1.86) | 2.02 (1.94-2.11) |
| Medication use | |||
| Antidepressants | 1.23 (1.21-1.25) | 1.25 (1.20-1.30) | 1.34 (1.28-1.41) |
| Benzodiazepines | 1.14 (1.10-1.17) | 1.13 (1.07-1.18) | 1.06 (1.00-1.13) |
| Antihypertensives | 1.01 (1.00-1.03) | 1.02 (1.00-1.05) | 1.02 (0.98-1.05) |
| Antidiabetes drugs | 1.06 (1.04-1.09) | 1.02 (0.98-1.07) | 1.06 (1.00-1.11) |
| Insulin | 1.45 (1.39-1.51) | 1.35 (1.24-1.46) | 1.52 (1.39-1.66) |
| ACB-2 drugs | 1.51 (1.42-1.61) | 1.43 (1.28-1.60) | 1.61 (1.40-1.84) |
| ACB-3 drugs | 1.09 (1.07-1.11) | 1.03 (0.99-1.08) | 1.07 (1.02-1.12) |
| Systemic steroids | 1.27 (1.24-1.30) | 1.20 (1.15-1.25) | 1.30 (1.24-1.37) |
Abbreviations: ACB, anticholinergic burden; AMD, age-related macular degeneration; CCI, Charlson Comorbidity Index; HR, hazard ratio; IMD, index of multiple deprivation; NA, not applicable.
As cases and controls were matched on age and sex, there were no differences between the 2 groups in the Cox models.
Race categories were compared with White, IMD compared with 1 (most deprived), and CCI score compared with 0.
Figure 2. Multivariable Hazard Ratios (HRs) for Falls, Fractures, and Body Site–Specific Fractures.
Multivariable-adjusted HRs for primary and secondary outcomes in individuals with cataract, age-related macular degeneration (AMD), or glaucoma compared with matched comparators without eye disease.
Results of our sensitivity analyses are presented in eTables 4 and 5 in Supplement 1. The inverse probability treatment weight models showed very similar HRs compared with standard Cox models. Further increased HRs were observed for both falls and fractures for all cases with single eye disease (ie, without history of concurrent eye disease) compared with comparators. For falls, the risks in the cataract-only group showed an HR of 1.44 (95% CI, 1.43-1.46); for AMD only, an HR of 1.91 (95% CI, 1.78-2.05); and for glaucoma only, an HR of 2.40 (95% CI, 2.28-2.54). For fractures, the risks for cataract only were an HR of 1.36 (95% CI, 1.33-1.38); for AMD only, an HR of 1.72 (95% CI, 1.57 - 1.88); and for glaucoma only, an HR of 2.17 (95% CI, 2.02-2.33).
Incident Fractures by Body Site
The results of the multivariable Cox regression analysis for body site–specific fractures are shown in Figure 2 and eTable 6 in Supplement 1. Overall, all populations with eye disease had an increased risk of fractures of almost all body sites (including hip, spine, forearm, skull or facial bones, pelvis, ribs or sternum, and lower leg fractures) compared with comparators. The exceptions were for hip, pelvic, and skull and facial bone fractures in the glaucoma cohort and rib and sternal fractures for the AMD cohort, where no differences were found.
Discussion
In this large population-based cohort study, we observed an increased risk of both falls and fractures for people with cataract, AMD, or glaucoma. This took into account a higher level of comorbidity in the population with eye disease, including multiple long-term conditions and increased medication usage. The highest effect sizes of covariates observed were higher levels of comorbidity, though not the extremes (ie, Charlson Comorbidity Index ≥12), or a history of fall or fracture. Site-specific analyses revealed an increased risk of almost all body sites, with particularly high risks for forearm and lower leg fractures. Our findings further build the evidence base demonstrating that all 3 eye diseases are important risk factors for falls and fractures. Our sensitivity analyses reported a further elevated risk within subgroup analyses of participants with single eye diseases only. Correspondingly, we found a higher risk of falls resulting in injury and have demonstrated an increased risk of both higher-impact (eg, skull and facial bones or pelvic fractures) and lower-impact fractures (eg, spinal fractures) via site-specific analyses. These findings contribute observational evidence supporting higher risks of falls or injuries for these populations and suggests a need to assess the medical and rehabilitation needs of at-risk individuals in future research.
Our findings contrast earlier cross-sectional studies reporting that only certain eye diseases are significant predictors of falls, although these all had self-reported falls outcomes that were potentially subject to recall bias.40 For example, a cross-sectional survey including 3280 older adults in East Asia reported a 4-fold increase in odds of falling in those with glaucoma, 1.5 times for cataract, but only 0.3 times for AMD. However, other small observation studies in France and Canada have found almost twice the risk of injurious falls in people with AMD.41,42 In practice, our HRs represent an increased risk over the study duration (median follow-up was approximately 4 years) but does not guarantee that the relative risk remains constant over this time. Further evidence is still needed examining the subsequent effect of increased risk of falls and fractures, potentially through examining linked outcomes such as related hospitalization and reductions in quality of life or quantifying financial effects through economic analyses. This may help further define which patients are particularly higher risk and need to access fall services and treatments more urgently.
Limitations
Although drawing from routine electronic health data allowed a large sample size, its retrospective nature is limited by imperfect data relying on coding and irregular follow-up. There is likely a small proportion of misclassification bias, as patients who have eye disease but have yet to be assessed by an ophthalmologist (eg, early cataracts or waiting lists) may have been misclassified as not having disease. Also, we were unable to assess visual function objectively within the analysis or whether the diagnosis was monocular or binocular. Furthermore, a key limitation is that we were unable to examine treatments during follow-up because of potential inaccuracies in coding, such as cataract surgery or medication for AMD or glaucoma, which may have overestimated our HRs. This is particularly the case for the cataract cohort, where surgery can restore normal vision promptly, with cases having artificially longer follow-up. Yet this may still carry subsequent risks, including posttreatment risks such as spectacle imbalance and posterior capsule opacification. For these reasons, our analyses may be more reflective of the typical patient experience, as we theorized that a diagnosis of any eye disease may be itself a marker of increased risk.
Previous studies have reported good validity in the reporting of fractures, but there remains potential biases for our outcomes.43,44,45,46 As we captured all-cause traumatic fractures, a proportion may have not related to falls, but we were unable to ascertain whether this was the case. Despite this, our analysis for fractures is likely to be more accurate than for falls because of the multifactorial nature of falls. First, there may be an underestimation of falls risk as people may only present having sustained a serious fall, with injurious falls more likely to be coded. Second, other factors affecting exposure to falls risk, such as cognitive status and physical activity, were not explicitly considered, although dementia and multiple comorbidities were further adjusted for within the inverse proportional treatment weight models, which indicated very similar hazard ratios. Younger, more physically active people may have a greater risk, and though increasing comorbidity generally increases risk, very high levels of comorbidity may actually limit activity as may be observed implicitly in our reduced effect sizes of extreme Charlson Comorbidity Index scores.
Conclusions
The results of this study indicate that people with cataract, AMD, or glaucoma have a higher risk of both falls or fractures compared with people without these eye diseases. These populations would likely benefit from improved advice, access, and referrals to falls prevention services and targeted interventions to prevent related adverse outcomes.
eTable 1. Codes for identifying eye diseases
eTable 2. Proportions and crude rates of primary and secondary outcomes
eTable 3. Incidence rates of falls and fractures (age standardized per 100,000 person years)
eTable 4. Sensitivity analyses results tables (inverse proportional treatment weight models
eTable 5. Sensitivity analysis 2 (single eye disease subanalysis)
eTable 6. Multivariate-adjusted hazard ratios for incident fractures by body site with covariates
eTable 7. Additional analysis (CPRD Aurum only analyses)
Data sharing statement
References
- 1.Bourne R, Steinmetz JD, Flaxman S, et al. ; GBD 2019 Blindness and Vision Impairment Collaborators; Vision Loss Expert Group of the Global Burden of Disease Study . Trends in prevalence of blindness and distance and near vision impairment over 30 years: an analysis for the Global Burden of Disease Study. Lancet Glob Health. 2021;9(2):e130-e143. doi: 10.1016/S2214-109X(20)30425-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Burton MJ, Ramke J, Marques AP, et al. The Lancet Global Health Commission on Global Eye Health: vision beyond 2020. Lancet Glob Health. 2021;9(4):e489-e551. doi: 10.1016/S2214-109X(20)30488-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.E JY, Li T, McInally L, et al. Environmental and behavioural interventions for reducing physical activity limitation and preventing falls in older people with visual impairment. Cochrane Database Syst Rev. 2020;9(9):CD009233. doi: 10.1002/14651858.CD009233.pub3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Boyd M, Kvizhinadze G, Kho A, Wilson G, Wilson N. Cataract surgery for falls prevention and improving vision: modelling the health gain, health system costs and cost-effectiveness in a high-income country. Inj Prev. 2020;26(4):302-309. doi: 10.1136/injuryprev-2019-043184 [DOI] [PubMed] [Google Scholar]
- 5.GBD 2017 Disease and Injury Incidence and Prevalence Collaborators . Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392(10159):1789-1858. doi: 10.1016/S0140-6736(18)32279-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.World Health Organization . Step safely: strategies for preventing and managing falls across the life-course. 2021. Accessed November 21, 2023. https://apps.who.int/iris/rest/bitstreams/1343716/retrieve
- 7.Davis JC, Robertson MC, Ashe MC, Liu-Ambrose T, Khan KM, Marra CA. International comparison of cost of falls in older adults living in the community: a systematic review. Osteoporos Int. 2010;21(8):1295-1306. doi: 10.1007/s00198-009-1162-0 [DOI] [PubMed] [Google Scholar]
- 8.Dhital A, Pey T, Stanford MR. Visual loss and falls: a review. Eye (Lond). 2010;24(9):1437-1446. doi: 10.1038/eye.2010.60 [DOI] [PubMed] [Google Scholar]
- 9.Reed-Jones RJ, Solis GR, Lawson KA, Loya AM, Cude-Islas D, Berger CS. Vision and falls: a multidisciplinary review of the contributions of visual impairment to falls among older adults. Maturitas. 2013;75(1):22-28. doi: 10.1016/j.maturitas.2013.01.019 [DOI] [PubMed] [Google Scholar]
- 10.Crabb DP. A view on glaucoma: are we seeing it clearly? Eye (Lond). 2016;30(2):304-313. doi: 10.1038/eye.2015.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kuang TM, Tsai SY, Hsu WM, Cheng CY, Liu JH, Chou P. Visual impairment and falls in the elderly: the Shihpai Eye Study. J Chin Med Assoc. 2008;71(9):467-472. doi: 10.1016/S1726-4901(08)70150-3 [DOI] [PubMed] [Google Scholar]
- 12.Lamoureux EL, Chong E, Wang JJ, et al. Visual impairment, causes of vision loss, and falls: the Singapore Malay Eye Study. Invest Ophthalmol Vis Sci. 2008;49(2):528-533. doi: 10.1167/iovs.07-1036 [DOI] [PubMed] [Google Scholar]
- 13.Wood JM, Lacherez P, Black AA, Cole MH, Boon MY, Kerr GK. Risk of falls, injurious falls, and other injuries resulting from visual impairment among older adults with age-related macular degeneration. Invest Ophthalmol Vis Sci. 2011;52(8):5088-5092. doi: 10.1167/iovs.10-6644 [DOI] [PubMed] [Google Scholar]
- 14.Bhorade AM, Perlmutter MS, Sabapathypillai SL, Goel M, Wilson B, Gordon MO. Rate of falls, fear of falling, and avoidance of activities at-risk for falls in older adults with glaucoma. Am J Ophthalmol. 2021;227:275-283. doi: 10.1016/j.ajo.2021.02.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hong T, Mitchell P, Burlutsky G, Samarawickrama C, Wang JJ. Visual impairment and the incidence of falls and fractures among older people: longitudinal findings from the Blue Mountains Eye Study. Invest Ophthalmol Vis Sci. 2014;55(11):7589-7593. doi: 10.1167/iovs.14-14262 [DOI] [PubMed] [Google Scholar]
- 16.Chu LW, Chi I, Chiu AY. Incidence and predictors of falls in the Chinese elderly. Ann Acad Med Singap. 2005;34(1):60-72. doi: 10.47102/annals-acadmedsg.V34N1p60 [DOI] [PubMed] [Google Scholar]
- 17.Javitt JC, Zhou Z, Willke RJ. Association between vision loss and higher medical care costs in Medicare beneficiaries costs are greater for those with progressive vision loss. Ophthalmology. 2007;114(2):238-245. doi: 10.1016/j.ophtha.2006.07.054 [DOI] [PubMed] [Google Scholar]
- 18.Bruce A. Visual impairment and the risk of mortality: addressing complex associations. J Epidemiol Community Health. 2016;70(8):746-747. doi: 10.1136/jech-2016-207580 [DOI] [PubMed] [Google Scholar]
- 19.Herrett E, Gallagher AM, Bhaskaran K, et al. Data resource profile: Clinical Practice Research Datalink (CPRD). Int J Epidemiol. 2015;44(3):827-836. doi: 10.1093/ije/dyv098 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wolf A, Dedman D, Campbell J, et al. Data resource profile: Clinical Practice Research Datalink (CPRD) Aurum. Int J Epidemiol. 2019;48(6):1740-1740g. doi: 10.1093/ije/dyz034 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Clinical Practice Research Datalink . CPRD data. Published 2022. Accessed May 1, 2023. https://cprd.com/data
- 22.McMillan B, Eastham R, Brown B, Fitton R, Dickinson D. Primary care patient records in the United Kingdom: past, present, and future research priorities. J Med Internet Res. 2018;20(12):e11293. doi: 10.2196/11293 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Domzaridou E, Carr MJ, Webb RT, Millar T, Ashcroft DM. Non-fatal overdose risk during and after opioid agonist treatment: a primary care cohort study with linked hospitalisation and mortality records. Lancet Reg Health Eur. 2022;22:100489. doi: 10.1016/j.lanepe.2022.100489 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Costello R, McDonagh J, Hyrich KL, Humphreys JH. Incidence and prevalence of juvenile idiopathic arthritis in the United Kingdom, 2000-2018: results from the Clinical Practice Research Datalink. Rheumatology (Oxford). 2022;61(6):2548-2554. doi: 10.1093/rheumatology/keab714 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kang EM, Pinheiro SP, Hammad TA, Abou-Ali A. Evaluating the validity of clinical codes to identify cataract and glaucoma in the UK Clinical Practice Research Datalink. Pharmacoepidemiol Drug Saf. 2015;24(1):38-44. doi: 10.1002/pds.3726 [DOI] [PubMed] [Google Scholar]
- 26.Nitsch D, Douglas I, Smeeth L, Fletcher A. Age-related macular degeneration and complement activation-related diseases: a population-based case-control study. Ophthalmology. 2008;115(11):1904-1910. doi: 10.1016/j.ophtha.2008.06.035 [DOI] [PubMed] [Google Scholar]
- 27.ClinicalCodes Repository . Clinical codes. Accessed November 21, 2023. https://clinicalcodes.rss.mhs.man.ac.uk/
- 28.National Institute for Health and Care Excellence (NICE) . Osteoporosis: assessing the risk of fragility fracture (CG146). 2018. Accessed November 21, 2023. https://www.nice.org.uk/guidance/cg146 [PubMed]
- 29.Curtis EM, van der Velde R, Moon RJ, et al. Epidemiology of fractures in the United Kingdom 1988-2012: variation with age, sex, geography, ethnicity and socioeconomic status. Bone. 2016;87:19-26. doi: 10.1016/j.bone.2016.03.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ravindrarajah R, Hazra NC, Charlton J, Jackson SHD, Dregan A, Gulliford MC. Incidence and mortality of fractures by frailty level over 80 years of age: cohort study using UK electronic health records. BMJ Open. 2018;8(1):e018836. doi: 10.1136/bmjopen-2017-018836 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Smith T, Noble M, Noble S, Wright G, McLennan D, Plunkett E. The English Indices of Deprivation 2015. London Dep Communities Local Gov; 2015. [Google Scholar]
- 32.Kontopantelis E, Mamas MA, van Marwijk H, et al. Geographical epidemiology of health and overall deprivation in England, its changes and persistence from 2004 to 2015: a longitudinal spatial population study. J Epidemiol Community Health. 2018;72(2):140-147. doi: 10.1136/jech-2017-209999 [DOI] [PubMed] [Google Scholar]
- 33.Bannay A, Chaignot C, Blotière PO, et al. The best use of the Charlson comorbidity index with electronic health care database to predict mortality. Med Care. 2016;54(2):188-194. doi: 10.1097/MLR.0000000000000471 [DOI] [PubMed] [Google Scholar]
- 34.Richardson K, Fox C, Maidment I, et al. Anticholinergic drugs and risk of dementia: case-control study. BMJ. 2018;361:k1315. doi: 10.1136/bmj.k1315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Hart LA, Phelan EA, Yi JY, Marcum ZA, Gray SL. Use of fall risk–increasing drugs around a fall-related injury in older adults: a systematic review. J Am Geriatr Soc. 2020;68(6):1334-1343. doi: 10.1111/jgs.16369 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Hippisley-Cox J, Coupland C. Derivation and validation of updated QFracture algorithm to predict risk of osteoporotic fracture in primary care in the United Kingdom: prospective open cohort study. BMJ. 2012;344:e3427. doi: 10.1136/bmj.e3427 [DOI] [PubMed] [Google Scholar]
- 37.Wang H, Abbas KM, Abbasifard M, et al. ; GBD 2019 Demographics Collaborators . Global age-sex-specific fertility, mortality, healthy life expectancy (HALE), and population estimates in 204 countries and territories, 1950-2019: a comprehensive demographic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1160-1203. doi: 10.1016/S0140-6736(20)30977-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Austin PC. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46(3):399-424. doi: 10.1080/00273171.2011.568786 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Brookhart MA, Schneeweiss S, Rothman KJ, Glynn RJ, Avorn J, Stürmer T. Variable selection for propensity score models. Am J Epidemiol. 2006;163(12):1149-1156. doi: 10.1093/aje/kwj149 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Ivers RQ, Cumming RG, Mitchell P, Attebo K. Visual impairment and falls in older adults: the Blue Mountains Eye Study. J Am Geriatr Soc. 1998;46(1):58-64. doi: 10.1111/j.1532-5415.1998.tb01014.x [DOI] [PubMed] [Google Scholar]
- 41.Radvay X, Duhoux S, Koenig-Supiot F, Vital-Durand F. Balance training and visual rehabilitation of age-related macular degeneration patients. J Vestib Res. 2007;17(4):183-193. doi: 10.3233/VES-2007-17404 [DOI] [PubMed] [Google Scholar]
- 42.Szabo SM, Janssen PA, Khan K, Lord SR, Potter MJ. Neovascular AMD: an overlooked risk factor for injurious falls. Osteoporos Int. 2010;21(5):855-862. doi: 10.1007/s00198-009-1025-8 [DOI] [PubMed] [Google Scholar]
- 43.Herrett E, Thomas SL, Schoonen WM, Smeeth L, Hall AJ. Validation and validity of diagnoses in the General Practice Research Database: a systematic review. Br J Clin Pharmacol. 2010;69(1):4-14. doi: 10.1111/j.1365-2125.2009.03537.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Van Staa TP, Abenhaim L, Cooper C, Zhang B, Leufkens HGM. The use of a large pharmacoepidemiological database to study exposure to oral corticosteroids and risk of fractures: validation of study population and results. Pharmacoepidemiol Drug Saf. 2000;9(5):359-366. doi: [DOI] [PubMed] [Google Scholar]
- 45.Österdahl MF, Sinnott SJ, Douglas I, Clegg A, Tomlinson L, Wong A. Frailty and rate of fractures in patients initiating antihypertensive medications: a cohort study in primary care. Drugs Ther Perspect. 2022;38(1):41-50. doi: 10.1007/s40267-021-00891-6 [DOI] [Google Scholar]
- 46.Archer L, Koshiaris C, Lay-Flurrie S, et al. ; STRAtifying Treatments In the multi-morbid Frail elderlY (STRATIFY) investigators . Development and external validation of a risk prediction model for falls in patients with an indication for antihypertensive treatment: retrospective cohort study. BMJ. 2022;379:e070918. doi: 10.1136/bmj-2022-070918 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Codes for identifying eye diseases
eTable 2. Proportions and crude rates of primary and secondary outcomes
eTable 3. Incidence rates of falls and fractures (age standardized per 100,000 person years)
eTable 4. Sensitivity analyses results tables (inverse proportional treatment weight models
eTable 5. Sensitivity analysis 2 (single eye disease subanalysis)
eTable 6. Multivariate-adjusted hazard ratios for incident fractures by body site with covariates
eTable 7. Additional analysis (CPRD Aurum only analyses)
Data sharing statement